84 



NORTH AMERICAN FAUNA 74 



With these additions, the formula of the walrus' deciduous dentition, 

 according to my findings, is 



. (l)-2-3 1 (l)-2-3-4 ^ o. . on 

 ^(iyT3 P^^x2 = 24to30. 



The teeth in parentheses were those found to be calcified in less than 50 % of 

 my specimens. The first upper and lower incisors and p^ were the latest to 

 develop and the least often calcified of any of the primary teeth, other than the 

 molars. 



In the secondary dentition, I identified four pairs of teeth in addition to those 

 of Cobb's formulae: P and the three lower incisors. Each of these also occurred 

 infrequently as a calcified tooth but usually was present as a distinct tooth germ 

 in the early developmental stages of the secondary dentition. By identification of 

 pi and its successor, I confirmed that the first permanent upper premolar in the 

 walrus, as in the northern fur seal (Kubota and Togawa 1970), is a tooth of the 

 secondary, rather than primary, dentition. 



The full comparative notation for the walrus' permanent dentition, with these 

 additions, is as follows: 



(l)-(2)- 3 1 l-2-3-(4) (l)-(2) ^ 2 - 18 -^8 



The teeth in parentheses were present in less than 50% of my adult specimens. 

 Thus, the walrus has the greatest potential number of teeth of any pinniped, but 

 the least number occurring with greater than 50% probability. Even with the 

 loss of Pj, the maximal formula of the walrus' permanent dentition is identical to 

 that of many living canoid carnivores. Presumably, this large number of teeth is 

 an atavistic character associated with the fissiped ancestry of the Otarioidea. 

 Each of the other genera of pinnipeds has typically two or three pairs of 

 deciduous and permanent incisors in the upper and one or two pairs in the lower 

 tooth rows. Such reductions from the primitive state (from three pairs in each 

 tooth row: Kellogg 1922) have been assumed to be due to loss of the first, rather 

 than last incisors (see Scheffer and Kraus 1964). That assumption is upheld by 

 my findings in the walrus. 



In the upper tooth row of the walrus, there are four pairs of permanent 

 premolars; in the lower row, three. Posterior to these in each row are one to two 

 pairs of smaller teeth, usually situated somewhat out of alignment with the 

 premolars. Cobb (1933) identified the latter as true molars (permanent teeth of 

 the primary dentition), and I have confirmed that. The fifth and sixth 

 postcanines in the upper row and the fourth and fifth in the lower row develop 

 from tooth germs of the primary dentition. Lingual primordia of secondary 

 successors occasionally are seen, but they apparently do not develop beyond that 

 stage. 



The four upper and three lower permanent premolars develop lingual to the 

 follicles of their primary antecedents. Hence, they are of the secondary dentition. 

 The first primary premolar (p^) in the upper row ordinarily does not develop 

 beyond the primordial stage, but p2-3-4 go on to become well-formed, calcified 

 teeth. The same pattern occurs in the northern fur seal (Kubota and Togawa 

 1970), in the harbor seal, Phoca vitulina (F. H. Fay and J. J. Burns, unpublished 



